Compare XPUs

Select up to 5 XPUs to compare side-by-side

Select XPUs to Compare

Clear all (3)Jump to results

Filter by Vendor

Showing 72 XPUs • 3 selected

Alibaba

Hanguang 800

AMD

MI100

23.1 TFLOPs

AMD

MI210

181 TFLOPs

AMD

MI250X

383 TFLOPs

AMD

MI300A

980.6 TFLOPs

AMD

MI300X

1,307 TFLOPs

AMD

MI325X

1,400 TFLOPs

AMD

MI350X

2,100 TFLOPs

AMD

Radeon PRO W7900

122 TFLOPs

AMD

Radeon RX 7900 XT

104 TFLOPs

AMD

Radeon RX 7900 XTX

122 TFLOPs

AWS

Inferentia2

190 TFLOPs

AWS

Trainium

190 TFLOPs

AWS

Trainium2

680 TFLOPs

Baidu

Kunlun II

Biren Technology

BR100

Cambricon

MLU370

256 TFLOPs

Cerebras

WSE-3

Enflame Technology

CloudBlazer T20

Etched

Sohu

10,000 TFLOPs

FuriosaAI

Warboy

Google

TPU v4

275 TFLOPs

Google

TPU v5e

197 TFLOPs

Google

TPU v5p

459 TFLOPs

Google

TPU v6e (Trillium)

918 TFLOPs

Graphcore

Bow IPU

Graphcore

IPU-M2000

Groq

LPU Inference Engine

Huawei

Ascend 910B

Iluvatar CoreX

BI-V150

300 TFLOPs

Intel

Data Center GPU Max 1100

177 TFLOPs

Intel

Data Center GPU Max 1550

419 TFLOPs

Intel Habana

Gaudi 2

432 TFLOPs

Intel Habana

Gaudi 3

1,835 TFLOPs

Multi-Metric Comparison

Relative performance across 5 key metrics (normalized to 100 = best in comparison)

Compute Performance (BF16)

Memory Capacity

Power Consumption

Power Efficiency

Specifications

Specification	NVIDIA L40S	NVIDIA B200	NVIDIA H100 SXM
Architecture	Ada Lovelace	Blackwell	Hopper
Form Factor	PCIe	SXM	SXM
VRAM	48 GB	192 GB	80 GB
Memory Bandwidth	864 GB/s	8,000 GB/s	3,350 GB/s
TFLOPs (FP32)	91.6	90	67
TFLOPs (FP16)	733	2,250	1,979
TFLOPs	733	2,250	1,979
TFLOPs (FP8)	1,466	4,500	3,958
TDP	350 W	1000 W	700 W
Launch Date	Oct 2023	Mar 2024	Sep 2022

Efficiency Metrics

Metric	L40S	B200	H100 SXM
TFLOPs per Watt (FP32-eq)	1.05	1.13	1.41
Memory Bandwidth per GB	18.0 GB/s	41.7 GB/s	41.9 GB/s

Performance Equivalence

How many units of each GPU are needed to match the performance of the others?

To match 1x NVIDIA L40S

NVIDIA B200

Compute (FP32-eq)

0.33x

B200 is 3.07x faster

FP32 Compute

1.02x

Need 1.02x B200

VRAM

0.25x

B200 has 4.00x more

Memory Bandwidth

0.11x

B200 has 9.26x more

NVIDIA H100 SXM

Compute (FP32-eq)

0.37x

H100 SXM is 2.70x faster

FP32 Compute

1.37x

Need 1.37x H100 SXM

VRAM

0.60x

H100 SXM has 1.67x more

Memory Bandwidth

0.26x

H100 SXM has 3.88x more

To match 1x NVIDIA B200

NVIDIA L40S

Compute (FP32-eq)

3.07x

Need 3.07x L40S

FP32 Compute

0.98x

L40S is 1.02x faster

VRAM

4.00x

Need 4.00x L40S

Memory Bandwidth

9.26x

Need 9.26x L40S

NVIDIA H100 SXM

Compute (FP32-eq)

1.14x

Need 1.14x H100 SXM

FP32 Compute

1.34x

Need 1.34x H100 SXM

VRAM

2.40x

Need 2.40x H100 SXM

Memory Bandwidth

2.39x

Need 2.39x H100 SXM

To match 1x NVIDIA H100 SXM

NVIDIA L40S

Compute (FP32-eq)

2.70x

Need 2.70x L40S

FP32 Compute

0.73x

L40S is 1.37x faster

VRAM

1.67x

Need 1.67x L40S

Memory Bandwidth

3.88x

Need 3.88x L40S

NVIDIA B200

Compute (FP32-eq)

0.88x

B200 is 1.14x faster

FP32 Compute

0.74x

B200 is 1.34x faster

VRAM

0.42x

B200 has 2.40x more

Memory Bandwidth

0.42x

B200 has 2.39x more

Pricing

Price Type	L40S	B200	H100 SXM
CAPEX (Street Price)	$10,000	$70,000	$30,000
OPEX (per hour)	$1.50/hr	—	$3.50/hr
Price per TFLOPs (FP32-eq)	$27	$62	$30